
\begin{table}[h!]
    \centering
    \caption{Developer Participation Responses to 421-a Incentive}
    \label{tab:developer_reponse}
    \small
    \begin{tabular*}{\textwidth}{@{\extracolsep{\fill}}lcccc@{}}
\toprule
& (1)             & (2)         & (3)       & (4)               \\ 
                & \multicolumn{2}{c}{LPM} & \multicolumn{2}{c}{Logit} \\ \cmidrule{2-3}\cmidrule{4-5}
                & Borough FE      & NTA FE   & Borough FE       & NTA FE    \\  \midrule  \addlinespace
421-a Incentive            &  0.557*** & 0.557***     & 5.184***     & 5.198***      \\
                &   (0.055) & (0.056)     & (0.538)     & 0.648)      \\ \addlinespace
$N$               & 11,669 & 11,666     & 11,669     & 7,465     \\
Clusters        & 179 & 176     & 179     & 82     \\
$R^2$ (within)     & 0.167 & 0.222      & n.a.         & n.a.       \\
Pseudo-$R^2$       & n.a.        & n.a.      & 0.233 & 0.292        \\  \addlinespace \midrule  \addlinespace
Marginal Effect &             &           & 0.183*** & 0.261***    \\
                &             &           & (0.033)     & 0.028)     \\  \addlinespace \bottomrule
\end{tabular*}
      

        	\begin{spacing}{1}
\begin{tablenotes}
      \item \footnotesize \textit{Notes:} This table reports estimates from Equation \ref{eq:logit} of the participation effects of the present-value tax-rate differential $\Delta\tau_i$ from providing onsite inclusionary housing under the 421-a exemption. All specifications include fixed effects for the borough and year of initial permit issuance. Marginal effects reflect the percentage-point increase in the onsite affordable housing participation rate per percentage-point increase in the tax incentive. Standard errors are clustered at the neighborhood level. $\sym{*} = p < 0.10$, $\sym{**} = p < 0.05$, $\sym{***} = p < 0.01$.
      \end{tablenotes}
       \end{spacing}
\end{table}